4 ms·

Human Connectome Project

36 points by jimsojim 11y ago

nulltype 11y ago
What level of detail is available from this data?
- josephdviviano 11y ago
  The resolution is some of the best in the business, so to speak, but the biological interpretation of all MRI-derived metrics are still hotly debated. So the true level is detail is sort of dependent on something we don't know yet.
  - nulltype 11y ago
    Hmm, is there something like this where they just slice a brain directly rather than using low resolution MRIs? Edit: oh here's one http://io9.com/see-the-first-ultra-high-resolution-3d-scan-of-the-ent-514395280 http://io9.com/see-the-first-ultra-high-resolution-3d-scan-o...
    - josephdviviano 11y ago
      Yes, but this does not include any functional data, arguably a huge contribution of the HCP project. The history of neuroscience has taught us that you can only look so far by looking at anatomy alone!
- fintler 11y ago
  http://www.humanconnectomeproject.org/data/inventory/ http://www.humanconnectomeproject.org/data/inventory/ If I'm reading it correctly, it looks like either 1.5mm, 2mm, or 2.5mm, depending on the dataset.
apl 11y ago
The key bottleneck with all connectome projects is still reconstruction. There's on the order of a hundred billion neurones in the human brain; they're connected via 1,000-10,000 times the number of synapses. Current machine learning approaches are decent, but still require massive amounts of proof-reading. Circuit neuroscience in Drosophila or the mammalian retina suggests neurones and synapses are indeed the level of detail we'll need in order to gain insight into computational mechanisms. So we're a long, long way off -- don't be too impressed by this particular dataset (which is insanely coarse!).
- SapphireSun 11y ago
  Hey if you guys want to help out with the proof reading effort, we've got a cool dataset that has a resolution of 16.5 nm x 16.5 nm x 23 nm at http://eyewire.org http://eyewire.org ^_^
fyrabanks 11y ago
This project and particular site have been around, largely unchanged, for many years. The WU-Minn side of the project is seemingly more active, and their dataset is much, much easier to obtain: http://humanconnectome.org/data/ http://humanconnectome.org/data/. Their scripts for processing the data are up on github at https://github.com/Washington-University/Pipelines https://github.com/Washington-University/Pipelines.

36 points by jimsojim 11y ago

nulltype 11y ago

What level of detail is available from this data?

josephdviviano 11y ago
The resolution is some of the best in the business, so to speak, but the biological interpretation of all MRI-derived metrics are still hotly debated. So the true level is detail is sort of dependent on something we don't know yet.
- nulltype 11y ago
  Hmm, is there something like this where they just slice a brain directly rather than using low resolution MRIs? Edit: oh here's one http://io9.com/see-the-first-ultra-high-resolution-3d-scan-of-the-ent-514395280 http://io9.com/see-the-first-ultra-high-resolution-3d-scan-o...
  - josephdviviano 11y ago
    Yes, but this does not include any functional data, arguably a huge contribution of the HCP project. The history of neuroscience has taught us that you can only look so far by looking at anatomy alone!
fintler 11y ago
http://www.humanconnectomeproject.org/data/inventory/ http://www.humanconnectomeproject.org/data/inventory/ If I'm reading it correctly, it looks like either 1.5mm, 2mm, or 2.5mm, depending on the dataset.

apl 11y ago

The key bottleneck with all connectome projects is still reconstruction. There's on the order of a hundred billion neurones in the human brain; they're connected via 1,000-10,000 times the number of synapses. Current machine learning approaches are decent, but still require massive amounts of proof-reading. Circuit neuroscience in Drosophila or the mammalian retina suggests neurones and synapses are indeed the level of detail we'll need in order to gain insight into computational mechanisms. So we're a long, long way off -- don't be too impressed by this particular dataset (which is insanely coarse!).

SapphireSun 11y ago
Hey if you guys want to help out with the proof reading effort, we've got a cool dataset that has a resolution of 16.5 nm x 16.5 nm x 23 nm at http://eyewire.org http://eyewire.org ^_^

fyrabanks 11y ago

This project and particular site have been around, largely unchanged, for many years. The WU-Minn side of the project is seemingly more active, and their dataset is much, much easier to obtain: http://humanconnectome.org/data/ http://humanconnectome.org/data/. Their scripts for processing the data are up on github at https://github.com/Washington-University/Pipelines https://github.com/Washington-University/Pipelines.